The present invention relates to an apparatus for presenting information for its analysis and evaluation.
Among the conventional technologies closely relating to the above type of apparatus are the numerical data visualizing technology, the knowledge expressing technology in the artificial intelligence, the hypertext browser technology, and the multiple variable data visualizing technology.
A graph generating function of table calculation software etc. is most typical of the numerical data visualizing technology. For example, in Microsoft Excel (trademark) of Microsoft Corp., numerical data that are input as a table can be expressed as a graph such as a scatter diagram (Reference (1)). The numerical data visualizing technology is also used to visualize scientific data. For example, Japanese Unexamined Patent Publication No. Hei. 4-229380 (Reference (2)) discloses conversion of numerical data to an image expressed as a hierarchical data structure. This image can be expressed at high speed as a three-dimensional or two-dimensional figure.
SemNet (Reference (3)) is known as the knowledge expressing technology in the artificial intelligence. To present to a user the status of a knowledge-base having a large amount of knowledge, knowledge is arranged in a three-dimensional space for presentation to a user. There are three methods of arranging knowledge in a three-dimensional space: arrangement according to attributes of respective pieces of knowledge, arrangement according to a relationship among pieces of knowledge, and arrangement according to a user's specification. In particular, in the arrangement according to attributes of respective pieces of knowledge, sequenced data received from a knowledge-base can be mapped to a three-dimensional space to enable display of nodes.
The hypertext browser technology includes Notecards (Reference (4)). NoteCards has a browser function of arranging, in a two-dimensional plane, icons that express information described on a card-by-card basis. Inclusion relationships, referencing relationships, etc. between cards are automatically expressed by lines. When an icon is indicated, the corresponding card is displayed to allow a user to recognize an information content of the card.
The multiple variable data visualizing technology includes S-MAP (Reference (5)), which enables display of multiple variable data on a single screen. For example, a relationship among customer data each consisting of the number of employees, sales, the number of business locations, the number of business people employed, etc. can be displayed on a single screen.
In each of the numerical data visualizing technology and the multiple data visualizing technology, numerical data are given in advance and visualization is performed based on those data. Therefore, these technologies cannot read and visually express data that cannot be dealt with as numerical data in their original form, for instance, language-information-based data. Even when indicating an image forming element representing each data, a user cannot know its content directly. A numerical data needs to be read from the position of a point or the like, or the corresponding location of the original data needs to be referred to. Further, in these technologies, the form of image forming elements cannot be changed in accordance with characteristics etc. of the information received.
In the knowledge expressing technology of SemNet, data that are expressed in a language are arranged in a space. However, to automatically arrange data in the space, it is necessary that a mechanism for sequencing the data be prepared and another data to be used for sequencing respective data be input in advance in a knowledge-base system other than SemNet. Therefore, SemNet itself cannot read and express three-dimensionally data that cannot be dealt with as numerical values. Further, even when indicating an image forming element representing each data, a user cannot know its content directly.
In the hypertext browser technology such as NoteCards, a relationship among pieces of information is determined automatically, and a user can know the content of an image called a node icon by mouse clicking. However, the method of arranging icons is limited to, for instance, using the relationship among pieces of information and specification by a user. Further, icons are not arranged in accordance with characteristics of respective pieces of information. Therefore, icons cannot be arranged in a space based on characteristics of respective data to allow a user to recognize the meaning of the whole icons.